Recent evidence obtained regarding the facultative methylotrophic bacterium, Methylobacterium extorquens AM1 suggests that methylotrophic metabolism in this organism represents an amalgam of metabolic routines formerly thought to be separated in different evolutionary branches. In order to address methylotrophy as a whole in this organism, it would be highly advantageous to carry out genome-level analysis. M. extorquens AM1 has a genome of 6 Mbp, and a complete sequencing project for this organism would be extremely expensive. However, physiological studies of this type do not require the complete genome sequence, assembled in a single contig. Instead, it is sufficient to obtain partial sequence of the majority of genes, suitable for database searches, insertional inactivation by PCR methods, and expression microarray-based analysis. We propose to carry out a "minimal genome sequencing project" in M. extorquens AM1 and use these data to further our understanding of methylotrophy. Random sequencing will be carried out to the 3X level, which should provide sequence data for about 95 percent of the genes. These data will be analyzed for genes of interest, and a subfraction of these will be chosen for targeted sequencing of adjacent areas. Genes of interest will be mutated and the phenotypes of the mutants will be analyzed. Expression array analysis will be carried out to identify genes that respond to common environmental cues, with emphasis placed on methylotrophy-related functions. The end result of this 4-year project will be the development of a metabolic framework within which our understanding of methylotrophy will be placed. The availability of sequence data for the majority of the genome will provide the basis for experimental approaches that should generate great leaps in our knowledge of this fundamental metabolic system. In addition, this project will provide a model for genome-level analysis of a number of microorganisms that represent the broad diversity of metabolism in the prokaryotic world, for which complete genome sequencing is unfeasible.